I am not a huge fan of Gartner or indeed any of the analysts. As a sweeping generalisation I feel that they take opinion and dress it up as fact (0.8 probability J) which I am uncomfortable with, especially when I don't agree with the opinion! I am also bored the whole magic quadrant thing which seems to me to be very speculative. However last week I was at a conference in the US where Tom Bittman, a Gartner VP and distinguished analyst (what a wonderful title) gave a talk on “Windows in a changing server market” which was truly excellent.
Tom’s premise that the server war (Windows vs Unix / Linux etc) was morphing into a infrastructure war (DSI vs On Demand). He had lots of interesting insights and facts to substantiate his hypothesis and then an analysis of what was important in the infrastructure space. He looked at security, virtualisation and skills in the infrastructure space (he missed one area that I think is key; organisational mapping) and then finally he looked at the suppliers and what they are doing, including Microsoft. He wasn’t very complementary about our efforts around DSI and had a number of suggestions about what we should be doing which I agreed totally with.
Anyway if you get a chance to hear Tom do go. He has an excellent grasp of the infrastructure space and doesn’t pull punches. He is also a good speaker.
I have to say that I find the whole magic box to be a bit of a stab in the dark, a qualitative analysis used as a quantative measure of feasibility. *whew* what a mouthful...but it's the truth, IMHO, that analysts are using increasingly more obscure methods of determining who and what the industry leaders are. Anyway, good point and insights into infrastructure, esp. the part about organizational (mind the z's here) structure and how it should map to a server environment.
I think one of the challenges in the DSI space are how does one vendor get another competing vendor with the same approach to create standard provisioning providers for the data center when they have no gain to do so.